Bringing our customers new solutions through digitalization
By adding robust new capabilities like machine learning to our microscopy systems, we are initiating a step-change in the way our customers in industry and academia manage and process vast amounts of imaging data generated by a range of imaging modalities. This enables them to easily and intelligently obtain scalable, quantitative insight.
The first algorithmic solution introduced by the ZEISS ZEN Intellesis platform makes integrated, easy to use, powerful segmentation for 2D and 3D datasets available to the routine microscopy user. ZEISS ZEN Intellesis software is available for the company’s full range of optical, confocal, X-ray, electron and ion microscopes using the ZEISS Efficient Navigation (ZEN) platform.
One of the principal challenges of microstructural imaging has been that these techniques are challenging to scale and automate, usually because the continuous outputs of the imaging techniques have to be ultimately classified into discrete phases for subsequent analysis and interpretation. These image outputs are subject to a variety of artifacts and noise that cause traditional analytical techniques to fail as the images become more complex.
During visual examination, the brain of a trained microscopist acts to integrate the rich, potentially multimodal datasets to extract the desired information. Such an approach is challenging to capture and express in a computational form, making microscopy challenging and expensive to scale across the many 1,000s of samples that may be required to upscale and contextualize research results. Machine learning techniques give us, for the first time, a powerful set of tools to capture the complex set of processes involved in analyzing the rich datasets available to microscopic imaging in a way that is computationally scalable to a much larger range of samples.
Firstly, machine learning based-classification schemes are much more noise tolerant than their traditional counterparts, and single high fidelity datasets can be used to train classifiers operating across wide areas of a sample. Even more exciting than this, machine learning can be used to discriminate features that have little or no difference in their greyscale values, but instead have differences that are discriminated by textural features alone. They can also be used to drive analysis based on spatially registered data integrated from multiple microstructural analysis techniques.
We have already explored and developed a range of different applications for machine learning technologies for geological, mineralogical and metals microstructural examination. Additional development is underway for life sciences, materials science and routine laboratory applications.
The post Introducing ZEISS ZEN Intellesis: Machine learning for microscopy appeared first on Microscopy News Blog.